Abstract Transcranial magnetic stimulation (TMS) is a non-invasive brain stimulation method which can alter brain activity in humans in a safe manner. Due its ease of application and ability to target specific brain regions, the repetitive application of TMS (rTMS) has the potential of augmenting or even replacing classic pharmacologic treatment-strategies. However, due to the enormous parameter space concerning its application (amplitude, coil position and orientation), optimal, i.e., personalized stimulation parameters for rTMS are very difficult to determine. Most importantly, principled efforts for optimizing TMS stimulation parameters are seriously limited by a poor understanding of molecular and cellular mechanisms. In-vitro recordings and functional optical imaging in slice preparations allow a direct assessment of the interaction of TMS electric fields with neurons in a highly controlled environment and allow for a rapid scan of stimulation parameters not achievable in human studies. Non-human primate (NHP) models are ideally suited to study circuit effects of TMS due their similarity to humans and the ability to perform invasive recordings to measure the electrophysiological response to TMS. Computational models can predict the electric field distribution of TMS in neural tissue, allowing to couple the biophysics of TMS with its physiological effects. Here, we propose to study the effects of changing TMS parameters in detail using in-vitro slice preparation, NHP recordings and computational modelling. We plan to study the molecular, cellular and circuit effects of transcranial magnetic stimulation in three specific aims: 1.) Systematic study of TMS stimulation parameters (stimulation intensity, coil orientation) in-vitro using mouse vs. rat hippocampal slices. 2) We will measure the circuit effects of varying TMS parameters in a NHP model. 3) Computational modelling of TMS electric fields across scales. Our findings will form the foundation for a mechanistic understanding between the biophysics of TMS electric fields and their physiological effects. This can form the basis for future efforts to improve noninvasive stimulation technologies.